Design Of A Fault Diagnosis System For High Voltage Circuit Breakers Based On FPGA

As the power equipment in the power system,high voltage circuit breaker plays an important role in the protection and control of the power grid.Its operation reliability is the key to the stable operation of power system.Therefore,the fault diagnosis of high voltage circuit breaker is of great significance.The vibration signal of the high-voltage circuit breaker carries a wealth of equipment status information,and the operation status of equipment can be diagnosed by extracting its vibration signal.Based on this,this paper will aim at developing a reasonable and available fault diagnosis platform for high voltage circuit breaker.Firstly,a Field-Programmable Gate Array(FPGA)chip is used as the control core to build a vibration signal acquisition system.And the appropriate hardware is selected,the driver is written in Verilog programming language,and the signal acquisition platform is built.ZW32-12 vacuum circuit breaker is taken as the research object,and the vibration signal in the four states of circuit breaker,such as normal,insufficient lubrication,base bolt loosening and energy storage spring falling off,is collected.The acquired vibration signal will be used for subsequent feature extraction and fault recognition.Secondly,the vibration signal is decomposed by Ensemble Empirical Mode Decomposition(EEMD),and the “sensitive” intrinsic mode function(IMF)component after signal decomposition is selected,and then the energy entropy value of the "sensitive" IMF component is calculated by using the energy entropy method.The energy entropy value is regarded as the eigenvectors of the vibration signal.Probabilistic neural network(PNN)is used as the circuit breaker fault identification method,and the obtained eigenvectors are input into PNN neural network for training and fault classification.And the diagnostic results are compared with support vector machine(SVM)and back propagation neural network(BP).From the comparison results,the recognition rate of PNN is higher and its advantages are more obvious.Finally,the fault diagnosis system of high voltage circuit breaker based on Graphical User Interface(GUI)of MATLAB is developed.Based on the theory of EEMD energy entropy-PNN and GUI support,a set of fault diagnosis system for high voltage circuit breaker is designed,which can be operated in real time without background programming.After the collected original vibration signal is input into the system,the system is operated with a simple mouse,and the obtained fault type is consistent with the actual result,which validates the rationality of the system effectively.